Arpita Ghosh

Unified Analysis of Secondary Traits in Case--Control Association Studies

It has been repeatedly shown that in case–control association studies, analysis of a secondary trait that ignores the original sampling scheme can produce highly biased risk estimates. Although a number of approaches have been proposed to properly analyze secondary traits, most approaches fail to reproduce the marginal logistic model assumed for the original case–control trait and/or do not allow for interaction between secondary trait and genotype marker on primary disease risk. In addition, the flexible handling of covariates remains challenging. We present a general retrospective likelihood framework to perform association testing for both binary and continuous secondary traits, which respects marginal models and incorporates the interaction term. We provide a computational algorithm, based on a reparameterized approximate profile likelihood, for obtaining the maximum likelihood (ML) estimate and its standard error for the genetic effect on secondary traits, in the presence of covariates. For completeness, we also present an alternative pseudo-likelihood method for handling covariates. We describe extensive simulations to evaluate the performance of the ML estimator in comparison with the pseudo-likelihood and other competing methods. Supplementary materials for this article are available online.

Glutathione S-transferase L1 multiplex serology as a measure of cumulative infection with human papillomavirus.

BackgroundSeveral assays are used to measure type-specific serological responses to human papillomavirus (HPV), including the bead-based glutathione S-transferase (GST)-L1 multiplex serology assay and virus-like particle (VLP)-based ELISA. We evaluated the high-throughput GST-L1, which is increasingly used in epidemiologic research, as a measure of cumulative HPV infection and future immune protection among HPV-unvaccinated women.MethodsWe tested enrollment sera from participants in the control arm of the Costa Rica Vaccine Trial (nu2009=u2009488) for HPV16 and HPV18 using GST-L1, VLP-ELISA, and two assays that measure neutralizing antibodies (cLIA and SEAP-NA). With statistical adjustment for sampling, we compared GST-L1 serostatus to established HPV seropositivity correlates and incident cervical HPV infection using odds ratios. We further compared GST-L1 to VLP-ELISA using pair-wise agreement statistics and by defining alternate assay cutoffs.ResultsOdds of HPV16 GST-L1 seropositivity increased with enrollment age (ORu2009=u20091.20 per year, 95%CI 1.03-1.40) and lifetime number of sexual partners (ORu2009=u20092.06 per partner, 95%CI 1.49-2.83), with similar results for HPV18. GST-L1 seropositivity did not indicate protection from incident infection over 4xa0years of follow-up (HPV16 adjusted ORu2009=u20091.72, 95%CI 0.95-3.13; HPV18 adjusted ORu2009=u20090.38, 95%CI 0.12-1.23). Seroprevalence by GST-L1 (HPV16 and HPV18, respectively) was 5.0% and 5.2%, compared to 19.4% and 23.8% by VLP-ELISA, giving positive agreement of 39.2% and 20.8%. Lowering GST-L1 seropositivity cutoffs improved GST-L1/VLP-ELISA positive agreement to 68.6% (HPV16) and 61.5% (HPV18).ConclusionsOur data support GST-L1 as a marker of cumulative HPV infection, but not immune protection. At lower seropositivity cutoffs, GST-L1 better approximates VLP-ELISA.

An exposure-weighted score test for genetic associations integrating environmental risk factors

Current methods for detecting genetic associations lack full consideration of the background effects of environmental exposures. Recently proposed methods to account for environmental exposures have focused on logistic regressions with gene-environment interactions. In this report, we developed a test for genetic association, encompassing a broad range of risk models, including linear, logistic and probit, for specifying joint effects of genetic and environmental exposures. We obtained the test statistics by maximizing over a class of score tests, each of which involves modified standard tests of genetic association through a weight function. This weight function reflects the potential heterogeneity of the genetic effects by levels of environmental exposures under a particular model. Simulation studies demonstrate the robust power of these methods for detecting genetic associations under a wide range of scenarios. Applications of these methods are further illustrated using data from genome-wide association studies of type 2 diabetes with body mass index and of lung cancer risk with smoking.

Demand- and supply-side determinants of diphtheria-pertussis-tetanus nonvaccination and dropout in rural India

Background Although 93% of 12- to 23-month-old children in India receive at least one vaccine, typically Bacillus Calmette–Guérin, only 75% complete the recommended three doses of diphtheria-pertussis-tetanus (DPT, also referred to as DTP) vaccine. Determinants can be different for nonvaccination and dropout but have not been examined in earlier studies. We use the three-dose DPT series as a proxy for the full sequence of recommended childhood vaccines and examine the determinants of DPT nonvaccination and dropout between doses 1 and 3. Methods We analyzed data on 75,728 6- to 23-month-old children in villages across India to study demand- and supply-side factors determining nonvaccination with DPT and dropout between DPT doses 1 and 3, using a multilevel approach. Data come from the District Level Household and Facility Survey 3 (2007–08). Results Individual- and household-level factors were associated with both DPT nonvaccination and dropout between doses 1 and 3. Children whose mothers had no schooling were 2.3 times more likely not to receive any DPT vaccination and 1.5 times more likely to drop out between DPT doses 1 and 3, compared with children whose mothers had 10 or more years of schooling. Although supply-side factors related to availability of public health facilities and immunization-related health workers in villages were not correlated with dropout between DPT doses 1 and 3, children in districts where 46% or more villages had a healthcare subcentre were 1.5 times more likely to receive at least one dose of DPT vaccine compared with children in districts where 30% or fewer villages had subcentres. Conclusions Nonvaccination with DPT in India is influenced by village- and district-level contextual factors over and above individuals’ background characteristics. Dropout between DPT doses 1 and 3 is associated more strongly with demand-side factors than with village- and district-level supply-side factors.

Quantitative estimates of dietary intake with special emphasis on snacking pattern and nutritional status of free living adults in urban slums of Delhi: impact of nutrition transition

BackgroundThe nutritional landscape of India is experiencing the fallout of urbanization and globalization. The changes are manifest in dietary patterns as well as health outcomes. The study aimed at assessing household dietary intake pattern with special emphasis on snacking pattern, anthropometric and lipid profiles in low socio-economic status households in an urban slum of Delhi.MethodsCommunity based cross-sectional study in 260 households of a purposively selected urban slum in North-East district of Delhi, India. Family dietary surveys including consumption pattern of commercial food products rich in Partially Hydrogenated Vegetable Oils (PHVOs), 24xa0h dietary recall and assessment of dietary diversity using Household Diet Diversity Scores (HDDS) were done. Assessment of nutritional status using anthropometric and lipid profile on a subsample (n =130) were also conducted.ResultsMedian energy and fat intake were adequate. Micronutrient intake was found to be inadequate for vitamin A, riboflavin, calcium and folate. PHVO usage was low (<20xa0% households). Milk (39xa0%), green leafy vegetables (25xa0%) and fruits (25xa0%) intake were below recommendations. Mean HDDS was 7.87. Prevalence of overweight/obesity was high (66.7xa0%). Lipid profile showed mean HDL-C levels lower than recommendations for females.ConclusionCommunity based awareness programs for prevention of non-communicable diseases should incorporate healthy diet and lifestyle practices with emphasis on quantity and quality of nutrient intake. This must be considered as an integral part of chronic disease prevention strategy for underprivileged communities in urban India.

Determinants of cost of routine immunization programme in India

The costs of delivering routine immunization services in India vary widely across facilities, districts and states. Understanding the factors influencing this cost variation could help predict future immunization costs and suggest approaches for improving the efficiency of service provision. We examined determinants of facility cost for immunization services based on a nationally representative sample of sub-centres and primary health centres (99 and 89 facilities, respectively) by regressing logged total facility costs, both including and excluding vaccine cost, against several explanatory variables. We used a multi-level regression model to account for the multi-stage sampling design, including state- and district-level random effects. We found that facility costs were significantly associated with total doses administered, type of facility, salary of the main vaccinator, number of immunization sessions, and the distance of the facility from the nearest cold chain point. Use of pentavalent vaccine by the state was an important determinant of total facility cost including vaccine cost. India is introducing several new vaccines including some supported by Gavi. Therefore, the government will have to ensure that additional resources will be made available after the support from Gavi ceases.

Productive disruption: opportunities and challenges for innovation in infectious disease surveillance

### Summary boxnnInfectious diseases place an unacceptable and disproportionate social and economic burden on low-income countries. National disease control programmes have the difficult task of allocating limited budgets for interventions across regions of their countries, based on often disparate datasets of varying quality from a range of sources including clinics, hospitals, village health workers, the private sector and non-governmental organisations (NGOs). Every stage of the data collection and analysis pipeline for surveillance systems may be affected by a lack of capacity as well as by biases and misaligned incentives for reporting and managing data. Addressing these issues will bexa0essential for effective reduction in the burden of endemic infectious diseases globally as well as to preparing for emerging epidemic threats.nnMeanwhile, academic researchers—often in high-income settings—are developing increasingly sophisticated methods to collect and analyse data to improve spatial estimates of disease burden using new Big Data sources, mobile-Health or m-Health approaches or mechanistic and statistical modelling techniques. While these advances leap ahead, however, many remain most useful for estimating global disease distribution,1 rather than for national control programme prioritisation. Translating these new techniques …

Neighborhood heterogeneity in health and well-being among the elderly in India – Evidence from Study on global AGEing and adult health (SAGE)

ABSTRACT We establish a rationale for a multilevel approach in examining health among older adults. Using data on a nationally representative sample of 6560 Indian adults aged 50 years and older, we examine the extent of contextual variation between neighborhoods, after accounting for the compositional effect of individuals’ background characteristics, across multiple dimensions of elderly health. The variance apportioned to neighborhoods in null intercept‐only models varied widely across different health outcomes examined in the elderly – while neighborhoods accounted for only 4% of the total variation in high blood pressure at exam, 23% of the total variation in self‐rated poor quality of life could be attributed to neighborhood‐level differences. In models that accounted for state, place of residence, and demographic and socioeconomic characteristics of individuals, the contribution of neighborhood to the total variation for most health outcomes was attenuated (2–11%) but persisted to exist. Our findings underscore the importance of neighborhoods in studying the health and well‐being of the elderly in India. HighlightsHealth and well‐being of the Indian elderly varies markedly across neighborhoods.Extent of neighborhood variation differs across dimensions of elderly health.Neighborhood variation persists after accounting for individual characteristics.Study suggests the importance of neighborhoods for late‐life health in India.

Leveraging Family History in Population‐Based Case‐Control Association Studies

Population‐based epidemiologic studies often gather information from study participants on disease history among their family members. Although investigators widely recognize that family history will be associated with genotypes of the participants at disease susceptibility loci, they commonly ignore such information in primary genetic association analyses. In this report, we propose a simple approach to association testing by incorporating family history information as a “phenotype.” We account for the expected attenuation in strength of association of the genotype of study participants with family history under Mendelian transmission. The proposed analysis can be performed using standard statistical software adopting either a meta‐ or pooled‐analysis framework. Re‐analysis of a total of 115 known susceptibility single‐nucleotide polymorphisms, discovered through genome‐wide association studies for several disease traits, indicates that incorporation of family history information can increase efficiency by as much as 40%. Efficiency gain depends on the type of design used for conducting the primary study, extent of family history, and accuracy and completeness of reporting.

Impact evaluation of a community engagement intervention in improving childhood immunization coverage: a cluster randomized controlled trial in Assam, India

BackgroundTo improve immunization coverage, most interventions that are part of the national immunization program in India address supply-side challenges. But, there is growing evidence that addressing demand-side factors can potentially contribute to improvement in childhood vaccination coverage in low- and middle-income countries. Participatory engagement of communities can address demand-side barriers while also mobilizing the community to advocate for better service delivery. The objective of this study is to evaluate the impact of a novel community engagement approach in improving immunization coverage. In our proposed intervention, we go a step beyond merely engaging the community and strive towards increasing ‘ownership’ by the communities.Methods/DesignWe adopt a cluster randomized design with two groups to evaluate the intervention in Assam, a state in the northeast region of India. To recruit villages and participants at baseline, we used a two-stage stratified random sampling method. We stratified villages; our unit of randomization, based on census data and randomly selected villages from each of the four strata. At the second-stage, we selected random sub-sample of eligible households (having children in the age group of 6–23xa0months) from each selected village. The study uses a repeated cross sectional design where we track the same sampled villages but draw independent random samples of households at baseline and endline. Total number of villages required for the study is 180 with 15 eligible HHs from each village. Post-baseline survey, we adopt a stratified randomization strategy to achieve better balance in intervention and control groups, leveraging information from the extensive baseline survey.DiscussionThe proposed intervention can help identify barriers to vaccination at the local level and potentially lead to more sustainable solutions over the long term. Our sampling design, sample size calculation, and randomization strategy address internal validity of our evaluation design. We believe that it would allow us to causally relate any observed changes in immunization coverage to the intervention.Trial registrationThe trial has been registered on 7th February, 2017 under the Clinical Trials Registry- India (CTRI), hosted at the ICMR’s National Institute of Medical Statistics, having registration number CTRI/2017/02/007792. This is the original study protocol.